A/B Test: uppercase vs lowercase in digital ads
Always be testing.
We must test our assumptions. Where there is existing and good quality research, use that. Test for what we don’t know and can’t discover from existing research.
I never want to be complacent—I am always looking for opportunities to discover whether there are ways to better engage our audience and get better results.
Backstory
When this A/B test was conducted, I was responsible for digital ad creative for a DTC CPG startup. We were running short motion graphics video ads across social media and display.
I am a UX-informed brand person and I espouse the belief that inclusive design benefits everyone, not just the audience these alterations have been designed for. Examples:
Wheelchair ramps and drop-curbs help wheelchair users. They also help those who are loaded down with heavy bags, strollers, rolling suitcases, or those impaired by a temporary injury.
Closed captions were invented for d/Deaf and hard-of-hearing people. They also help people watch videos next to their partner or sleeping child in bed. They also help people watch videos in public places discreetly or watch videos while listening to music.
Assistive touch on devices was invented for those with motor disabilities and it’s now being used to give people the ability to command TikTok to advance to the next video without touching the phone (great for watching TikTok while doing dishes or ironing!)
Lowercase lettering was devised as a fast(er) way for medieval scribes to reproduce manuscripts that could take months to copy by hand. It so happens that lowercase lettering is also faster for the reader (33% faster in fact). This means that parents with screaming toddlers in a retail environment can more quickly scan products on the shelves, understand the claims on the packaging, and make a purchasing choice.
Hypothesis
My hypothesis was that audiences would:
Abandon ads delivered in all uppercase lettering
Spend longer watching ads delivered in lowercase lettering and, thereby, be more likely to convert.
Experiment
For two weeks, we ran two variations of the same video as ads on Facebook. One with text in all block caps and one with all lowercase.
Results
The lowercase version did 500% better for viewers watching 100% of the ad and converted into sales at a 600% higher rate.
For this test, the numbers are low. If scaled, they would be significant to the ad’s ROI.
What’s Next
What do we do with this information? For me, it goes into the vault of research that I lean on to build my understanding of the best practices, not just for ad design, but design in general.
